- Email: [email protected]
Technique effects on upper limb loading in the tennis serve
Technique effects on upper limb loading in the tennis serve B Elliott 1, G Fleisig 2, R Nicholls I & R Escamilia 3 1The School of Human Movement and Exercise Science, The University of Western Australia, Australia. 2American Sports Medicine Institute, Birmingham, Alabama, USA. 3California State University, Sacramento, California, USA.
Elliott, B, Fleisig, G, Nicholls, R, & EscamiUa, R (2003). Technique effects on upper limb loading in the tennis serve. Journal of Science and Medicine in Sport 6 (1): 76-87. The purpose of this study was to compare the shoulder and elbow joint loads during the tennis serve. Two synchronised 200 Hz video cameras were used to record the service action of 20 male and female players at the Sydney 2000 Olympics. The displacement histories of 20 selected landmarks, were calculated using the direct linear transformation approach. Ball speed was recorded from the stadium radar gun. The Peak Motus system was used to smooth displacements, while a customised inverse-dynamics program was used to calculate 3D shoulder and elbow joint kinematics and kinetics. Male players, who recorded significantly higher service speeds (male=183 km hr-l: female=149 km hr -1) recorded significantly higher normalised and absolute internal rotation shoulder torque at the position when the arm was maximally externally rotated (MER) (male=4.6% and 64.9 Nm: female=3.5% and 37.5 Nm). A higher absolute elbow varus torque (67.6 Nm) was also recorded at MER, when compared with the female players (41.3 Nm). Peak normalised horizontal adduction torque (male=7.6%: female=6.5%), normalised shoulder compressive force (male=79.6%: female=59.1%) and absolute compressive force (male=608.3 N: female=363.7 N}, were higher for the male players. Players who flexed at the front knee by 7.6 °, in the backswing phase of the serve, recorded a similar speed (162 km hr-1), and an increased normalised internal rotation torque at MER (5.0%), when compared with those who flexed by 14.7 °. They also recorded a larger normalised varus torque at MER (5,3% v 3.9%) and peak value (6.3% v 5.2%). Players who recorded a larger knee flexion also recorded less normalised and absolute (4.3%, 55.6 Nm) peak internal rotation torque compared with those with less flexion {5.6%, 63.9 Nm). Those players who used an abbreviated backswing were able to serve with a similar speed and recorded similar kinetic values. Loading on the shoulder and elbow joints is higher for the male than female players, which is a reason for the significantly higher service speed by the males, The higher kinetic measures for the group with the lower knee flexion means that all players should be encouraged to flex their knees during the backswing phase of the service action. The type of backswing was shown to have minimal influence on service velocity or loading of the shoulder and elbow joints.
Introduction T h r e e - d i m e n s i o n a l (3D) k i n e m a t i c a n a l y s e s of t h e service t e c h n i q u e of h i g h performance tennis players have provided the athlete a n d coach with practical i n f o r m a t i o n o n t h e key m e c h a n i c a l c h a r a c t e r i s t i c s of t h i s a c t i o n (1,2). However, s p o r t p h y s i c i a n s , t h e r a p i s t s , a n d t r a i n e r s n e e d to k n o w t h e forces a s s o c i a t e d
76
Technique effects on upper limb loading in the tennis serve
with these movements, as the aetiotogy of musculoskeletal injuries generally h a s a kinetic mechanical c a u s e (a). A better u n d e r s t a n d i n g of the physical d e m a n d s placed u p o n the body during this activity will enable health professionals to design more optimal injury prevention, t r e a t m e n t a n d rehabilitation programs. While epidemiological studies have reported a range of injuries in tennis, elite players c o m m o n l y suffer from shoulder a n d medial elbow injuries (4, 5). Hill (6} reported 56% of competitive players suffered rotator cuff and i m p i n g e m e n t injuries to the shoulder. It h a s b e e n h y p o t h e s i s e d t h a t the combination of a b d u c t i o n a n d external rotation of the u p p e r a r m overloads the static a n d dynamic stabilisers of the shoulder joint (7, 8). A high elbow v a r u s force is also p r e s e n t during the above actions, which h a s b e e n identified as the tennis stroke t h a t Causes m o s t medial elbow pain (9). In addition to these p r e p a r a t o r y m o v e m e n t s , the u p p e r a r m is involved in explosive inward rotation prior to ball impact, which m a y place additional s t r e s s on the shoulder region (2). While it h a s b e e n well d o c u m e n t e d t h a t the action of throwing in baseball is associated with high loads at the s h o u l d e r and elbow joints (1o), there is a paucity of information describing the loading of the shoulder and elbow joints during the tennis serve. B a h a m o n d e (111 a n d Noffal (12) in data presented as conference abstracts, reported p e a k shoulder internal rotation torques of approximately 50 Nm a n d 100 Nm respectively as the t r u n k rotates forward a n d the s h o u l d e r joint externally rotates. This occurs at the extreme limits of the shoulder joint range of motion, during the late stage of the service backswing a n d early forwardswing. They also reported a p e a k v a r u s torque at the elbow of approximately 43 Nm a n d 106 Nm respectively at this p h a s e of the service action. The u p p e r a r m in abducting, horizontally flexing a n d vigorously inwardly rotating to ball i m p a c t p r o d u c e d large shoulder horizontal flexion (164 Nm) a n d elbow v a r u s (74 Nm) torques (11). Dillman et al. (la) stated t h a t a n y torque greater t h a n 50 Nm in the u p p e r extremity was a significant factor in loading t h a t a r e a of the body. It is therefore a p p a r e n t that the u p p e r limb is subject to high loads during the service action, a n d this m o v e m e n t if repeated m a n y times m a y have the potential to c a u s e injury. Kibler (14, is) indicated t h a t a n y disruption to the kinetic chain could result in increased loading of other joints in the sequence of movements. While a coordinated action m a y reduce u p p e r limb loading in the serve, it is plausible t h a t modifications to this flow m a y increase forces on the shoulder and elbow joints. In baseball, changes to technique have been advocated in a n a t t e m p t to reduce forces a n d torques at the shoulder a n d elbow joints (16). Saal {5)stressed the need for a "deep knee bend" in the p r e p a r a t o r y tennis service m o v e m e n t s to avoid t h o r a c o l u m b a r hyperextension. Both a full backswing and a n effective "leg-drive" have b e e n advocated as features of a n efficient service technique (17) No r e s e a r c h h a s been published t h a t shows how modification to these actions affects shoulder and elbow joint loading. The p u r p o s e of this s t u d y w a s to c o m p a r e two variations in service technique for elite players u n d e r c h a m p i o n s h i p m a t c h conditions. Shoulder and elbow joint torques a n d forces at key p h a s e s of the service action were c o m p a r e d for players with a full and a n abbreviated backswing, a n d for those with a larger front knee joint flexion c o m p a r e d to those with minimal knee flexion. The above kinetic variables were also c o m p a r e d for male and female players. 77
Technique effects on upper limb loading in the tennis serve
Methods and Procedure All data were collected during singles matches on centre court during the XXVII Olympic Games in Sydney, Australia. Since professional tennis players are allowed in Olympic competition, most of the competitors were of professional status. Thirty-six tennis m a t c h e s were videotaped over a seven-day period with two electronically synchronised high-speed video cameras. H u m a n rights clearance was obtained from universities involved in the project and from the International Olympic Committee and the International Tennis Federation. Video data were collected at a rate of 200 Hz. For each camera, the s h u t t e r rate was set at 0.001 s and the aperture was adjusted according to weather conditions. Each camera was positioned approximately 20-40 m from the service area, with both side a n d front views. These views were adjusted to capture the entire serving motion, while optimising the field of view (Figure 1). Ball velocity was recorded as the ball left the tennis racket from a r a d a r g u n positioned in-line with the c e n t r e - m a r k and perpendicular to the baseline at the opposite end of the stadium. A 2 x l . 9 x l . 6 m calibration frame (Peak Performance Technologies, Inc., Englewood, CO), surveyed with a m e a s u r e m e n t tolerance of 0.005m, was videotaped at one end of the court prior to and after each match. The position of the frame e n c o m p a s s e d the space used by players during the serve (from initial ball toss to impact) on the ad-side (left-side) of one end of the court. Since kinematic parameters were m e a s u r e d only from ball toss to j u s t prior to ball impact, and not during the follow-through phase, the volume was large enough to contain the portion of the tennis serve that was digitised.
camer~
!!ii i
camerE
Figure 1: Camera locations on the centre court.
78
Technique effects on upper limb loading in the tennis serve
All videotapes were qualitatively analysed, a n d it was determined t h a t d a t a from 20 players were of the quality required for quantitative analysis. For each of these subjects, the three successful serves with the greatest ball velocity were m a n u a l l y digitised with the Peak Motus s y s t e m (Peak Performance Technologies, Inc., Englewood, CO). A 20-point spatial model was created, comprising the centres of the left a n d right mid-toes, ankles, knees, hips, shoulders, elbows, wrists, and the subject's head, h a n d racket grip, and four points a r o u n d the racket h e a d (medial, lateral, proximal, and distal). E a c h of these 20 points was digitised in every video field (200 Hz), starting at five f r a m e s prior to w h e n the ball left the h a n d a n d ending one frame prior to the racket impacting the ball. The Peak Motus s y s t e m t h e n calculated the 3D coordinate d a t a from the 2D digitised images utilising the direct linear t r a n s f o r m a t i o n method. An average root m e a n s q u a r e calibration error of 0.014 m was produced. A c o m p u t e r p r o g r a m was specifically written to calculate kinematic and kinetic p a r a m e t e r s . In this program, 3D coordinate d a t a were filtered with a 12 Hz low-pass Butterworth filter. Shoulder external rotation was calculated as the angle between the forearm and the anterior direction of the hittingshoulder, in a plane perpendicular to the long-axis of the u p p e r a r m (a horizontal f o r e a r m positioned anterior to the shoulder r e p r e s e n t s 0 °, while 90 ° of rotation h a s the forearm vertically aligned w h e n the u p p e r a r m is a b d u c t e d 90 ° to the trunk). Because external rotation is calculated using the forearm instead of the u p p e r arm, the a c c u r a c y of the calculation m a y diminish as the forearm a n d u p p e r a r m s e g m e n t s b e c o m e parallel. No m e a s u r e s were t a k e n at impact, w h e n this alignment is m o s t likely to occur. Flexion of the lead k n e e w a s determined as the relative angle between the thigh and the s h a n k (full knee extension equals 0 ° of flexion). Resultant force and torque levels at the elbow and shoulder joints were calculated u s i n g inverse dynamics. B e c a u s e of the dynamic effect of i m p a c t on the racket, kinetic analysis was limited to d a t a j u s t before ball impact. Drag effect of e a c h s e g m e n t t h r o u g h the air w a s neglected. Shoulder force w a s reported as the p r o x i m a l / d i s t a l and a n t e r i o r / p o s t e r i o r c o m p o n e n t s of the force applied by the t r u n k onto the u p p e r a r m (Figure 2A). Following a procedure first described b y Feltner and D a p e n a (is}, the s u m of all torques applied to each segment w a s set equal to the vector p r o d u c t of the s e g m e n t ' s m o m e n t of inertia a n d a n g u l a r acceleration. Moment of inertia of the racket a b o u t its medial-lateral axis w a s c o m p u t e d using the parallel axis t h e o r e m a n d published racket "swingweight" d a t a (19). Racket m o m e n t of inertia a b o u t the long-axis was calculated as: Moment of inertia (kgom 2) = Mass (kg)* [head width (m)]2/17.75 (20). Racket m o m e n t of inertia a b o u t its anteriorposterior axis w a s the s u m of the racket's other two principal m o m e n t s of inertia (20). Moment of inertia of the h a n d was a s s u m e d to be negligible. The positions of centre of m a s s of the f o r e a r m and u p p e r a r m were determined using cadaveric data (21). The m a s s of each u p p e r extremity s e g m e n t w a s a s s u m e d to be a percentage of the subject's total m a s s {211. Moments of inertia of the forearm a n d u p p e r a r m were initially determined from D e m p s t e r ' s (22) cadaveric study, and t h e n individualised using each subject's height a n d m a s s (2a). Shoulder internal rotation torque and horizontal adduction torque were the c o m p o n e n t s of the torque applied b y the t r u n k onto the u p p e r a r m
79
Technique effects on upper limb loading in the tennis serve
Anterior
\ Figure 2: Convention for kinetic measurements: (A) shoulder force, (B) shoulder torque, (C) elbow torque
Figure 3: Front knee joint flexion (~) in the service backswing (straight limb = 0 ° flexion).
Figure 4: Full (A) and abbreviated (B) backswings in the serve.
8O
Technique effects on upper limb loading in the tennis serve /
(Figure 2B). Elbow v a r u s torque was reported as the torque applied by the u p p e r a r m onto the forearm a b o u t the anterior direction of the f o r e a r m (Figure 2C). Elbow flexion torque was the torque applied b y the u p p e r a r m onto the forearm, a b o u t the medial direction of the elbow. Repeatability of data was a s s e s s e d by redigitising 30 consecutive frames, after one week, for the period from MER to immediately before impact. Shoulder internal rotation a n d elbow extension angles were c o m p a r e d using S t u d e n t ' s t-tests. Mean d a t a re-digitised over the 30 consecutive frames were shown to be reliable in t h a t shoulder internal rotation and elbow extension were not significantly different and varied by 0.8% a n d 0.4% respectively. Players were classified as exhibiting a n effective (>10 ° of knee flexion; Figure 3) or less-effective leg drive (<10 ° of knee flexion); at MER in the service action. Players were also categorised by their type of backswing. Players who u s e d a full rotation of the shoulder joint during the service action (Figure 4A) were classified as full backswing, w h e r e a s players who lifted the racket vertically during the early backswing (Figure 4B) were considered to have an abbreviated backswing. Players were also classified b y their gender. H y p o t h e s e s were that m a l e s would record higher kinetic values t h a n females, a n d those players with a n effective leg drive would record lower values t h a n t h o s e w i t h m i n i m a l k n e e flexion d u r i n g t h e b a c k s w i n g . It w a s also h y p o t h e s i s e d t h a t players with a full b a c k s w i n g would record lower kinetic values t h a n those with a n abbreviated backswing. All c o m p a r i s o n s were tested t h r o u g h a three way ANOVA (Data d e s k 6.1: D a t a Description Inc., USA). Ttests for u n e v e n cell sizes were u s e d to f u r t h e r c o m p a r e data where trends were identified from the above statistics (24). A 0.01 level of acceptance was established with respect to the above hypotheses. In addition to absolute values, n o r m a l i s e d values of all force and torque d a t a were calculated to improve c o m p a r i s o n of groups containing varying n u m b e r s of male a n d female players. Force d a t a were divided b y bodyweight and t h e n multiplied b y 100%, as h a s occurred in the baseball literature (16). Torque d a t a were divided by weight by height, t h e n multiplied b y 100% (16).
Results and DiScussion Gender, age, body mass, height, and service velocity for the m e a n of the b e s t three trials are s h o w n in Table 1. Serving speeds for the male players (182.8 k m h r -1) w a s significantly higher (F=24.7, p=0.01) t h a n for the female competitors (149.3 k m hr-]). The speed recorded b y both groups testifies to the elite quality of this sample, as values for high p e r f o r m a n c e d a t a in the literature are reported as approximately 155 a n d 125 k m hr -1 for male a n d female players respectively (1).
Male and female players Although the m e a n angle at MER w a s similar between groups (~170°), significantly larger normalised (F=5.4, p<0.01) a n d absolute (F= 12.2, p<0.01) internal rotation torques were recorded at this i n s t a n t for male (4.6%: 64.9 Nm) c o m p a r e d with female competitors (3.5%: 37.5 Nm) (Table 1). Similar m e a n angle (165 °) a n d absolute internal rotation torque (67 Nm) were recorded at MER for highly skilled male pitchers (16). While normalised m e a n elbow v a r u s torque at MER w a s similar for both groups, the absolute value was significantly higher for male players (67.6 Nm) t h a n for their female c o u n t e r p a r t s (41.3 Nm)
81
Technique effects on upper limb loading in the tennis serve
Anthropometric and ball velocity data
Males (n=8)
Females (n=12)
Height* (m) Mass* (kg) Velocity* (km hr1)
185.0 (5.5) 81.0 (9.0) 182.8 (13,9)
174.0 (8.6) 62.2 (7.7) 149.3 (14,0)
Data at MER Shoulder internal rotation torque* (% weightxheight) Shoulder internal rotation torque* (Nm) Shoulder external rotation angle (°) Shoulder horizontal adduction torque (% weightxheight) Shoulder horizontal adduction torque (Nm) Elbow varus torque (% weightxheight) EIbowvarus torque* (Nm)
4.6 (0.9) 64.9 (15,8) 169,2 (9,0) 4,2 (1,7) 61.7 (31.0) 4.8 (0.9) 67.6 (15,2)
3,5 (1,2) 37,5 (15.0) 170.9 (8.1) 3.5 (1,4) 36,8 (16,3) 3,9 (1.2) 41.3 (14,1)
Peak values Shoulder internal rotation torque (% weightxheight) Shoulder internal rotation torque (Nm) Time prior to impact (s) Shoulder horizontal adduction torque* (% weightxheight) Shoulder horizontal adduction (Nm) Shoulder anterior force (% weight) Shoulder anterior force (N) Shoulder proximal force* (%weight) Shoulder proximal force* (N) Elbow flexion torque (%weightxheight) Elbow flexion torque (Nm) Elbow varus torque (% weightxheight) EIbowvarus torque** (Nm)
5.1 (0,9) 71.2 (15,1) 0.06 (0.02) 7.6 (0,8) 107.8 (24,9) 38.5 (14,0) 291.7 (119.8) 79.6 (5.3) 608.3 (109,5) 2.6 (1.6) 36.7 (22.9) 5.6 (0.9) 78.3 (12.2)
4.5 (1.3) 47.8 (16,3) 0,06 (0,02) 6,5 (0,9) 68,8 (14.3) 30.5 (10.2) 185.1 (60.9) 59.1 (8.4) 363.7 (87,8) 1,7 (1,3) 18.1 (14.5) 5.3 (1,1) 58.2 (13.1)
* Significant difference at the 0.01 level. **Significant difference at the 0.05 level as calculated using a t-test. Table 1:
Mean (Standard Deviation) service data for male and female players.
(F=13.2, p<0.01). Again data from the male players were almost identical to the 64 Nm reported by Fleisig et al. (16) for male baseball pitchers. The peak normalised horizontal adductor torque was significantly higher for the male players (7.6%)(F=7.47, p<0.01) when compared with the females (6.5%), while other shoulder torque values were similar in the swing to impact (Table 1). The m e a n horizontal adduction torque (107.8 Nm) for the male players was less t h a n reported by B a h a m o n d e (11) for college-level male tennis players (164 Nm), who p r e s u m a b l y served at a lesser speed t h a n this sample of professional players. Mean peak internal rotation torque values were relatively high during this phase of the service action for both groups (males = 71.2 Nm and females = 47.8 Nm), which supports the importance of internal rotation as one of the most important contributors to racket speed at impact (2). The level of internal rotation torque for the male players was larger t h a n the 33 Nm reported j u s t prior to impact by Bahamonde (11). However, it should be stressed that the values reported in this study were peak values, and not those immediately prior to impact. Normalised (F=6.7, (proximal) p<0.01) and absolute (F=7.3, p<0.01) m e a n peak shoulder compressive proximal forces were significantly larger for the 82
Technique effects on upper limb loading in the tennis serve
male players (79.6%: 608.3 N) c o m p a r e d with the female competitors (59.1%: 363.7 N). This is again similar to baseball (16). To prevent distraction at the shoulder d u n n g this p a r t of the action, the t r u n k applies a r e s u l t a n t force to the u p p e r a r m to stabilise the joint. Anterior forces a b o u t the shoulder joint were similar for all players irrespective of sex. The m a g n i t u d e s of shoulder kinetics during the p r e s e n t tennis s t u d y for the male players were smaller t h a n the shoulder kinetics previously reported for baseball pitching (16, 10) There w a s a trend for p e a k absolute elbow v a r u s torque to be higher for m a l e s (78.3 Nm) w h e n c o m p a r e d with the female players (58.2 Nm). The torque m a g n i t u d e for the male players was greater t h a n the 64 Nm reported for highly skilled a n d professional male baseball pitchers b y Fleisig et al. {16, 10), b u t less t h a n the 100 N m recorded by Feltner a n d D a p e n a t181 for baseball pitching. The fact t h a t the male players served at a significantly higher m e a n velocity was p r e s u m a b l y the result of the higher kinetic values in the u p p e r limb. The relationship between service velocity a n d u p p e r limb kinetics, particularly with respect to injury, is the subject of further investigation. Level Of knee flexion At MER there was a significantly lower n o r m a l i s e d internal rotation torque (3.5%) recorded by the group with a m e a n knee flexion of 15.9 °, w h e n c o m p a r e d with the group with a m e a n knee flexion 5.8 ° (5.0%: F= 8.08, p= 0.01). The group with the larger flexion recorded a similar level (43.7 Nm), a n d the less effective group a greater level (57.8 Nm) t h a n the 43 Nm reported by B a h a m o n d e (1~), for college level tennis players. However, both groups recorded lower values t h a n the 94 Nm reported by Noffal {12). This torque was associated with the eccentric contraction of the m u s c l e s responsible for internal shoulder rotation (eg latissimus dorsi, pectoralis m a j o r and subscapularis, teres major, and anterior deltoid) following MER. The players with the potential for a better leg-drive m a y therefore u s e the inertial transfer from the t r u n k to u p p e r limb to move the u p p e r a r m into a position of MER. This t h e n requires less internal rotator torque to stop the external rotation. The group with less effective drive m u s t therefore primarily use the external rotators to achieve MER, which t h e n requires a greater internal rotator torque to reverse the rotation of the u p p e r arm. A significantly reduced normalised (4.3 v 5.6%) a n d absolute p e a k internal rotation torque (55.6 Nm v 63.9 Nm), w a s recorded for the group with the larger knee joint flexion, w h e n c o m p a r e d with the less effective leg-flexion group. This reduction placed a lesser load on the joint during the concentric contraction of the m u s c l e s involved in rotating the u p p e r a r m in the swing to impact. The group with the larger knee flexion recorded a lower m e a n m a x i m u m internal rotator torque a n d the group with the lesser flexion a similar torque to the 68 Nm reported for professional pitchers (lo). Lesser load is t h u s evident with the group with the larger front knee flexion in performing an action described by Elliott et al. (2) as being integral to an effective service action. No significant differences were recorded w h e n the shoulder forces were c o m p a r e d between the two groups. A significantly larger m e a n elbow v a r u s torque was recorded at the position of MER for those players with a lesser knee flexion (5.3%: 60.2 Nm v 3.9%: 47.4 Nm). The group with the larger knee flexion recorded a similar v a r u s torque (43
83
Technique effects on upper limb loading in the tennis serve
Anthropometric and ball velocity data
Effective (9F & 5M)
Minimal (3F & 3M)
Height (m) Mass (kg) Service velocity (km hr "1) Knee flexion* (°)
178.3 (10.1) 69.0 (12.2) 162.9 (20.4) 15.9 (4.9)
178.5 66.5 162.2 5.8
(9.9) (11.0) (26.6) (1.8)
Data at MER Shoulder internal rotation torque* (% weightxheight) Shoulder internal rotation torque (Nm) Shoulder external rotation angle (°) Shoulder horizontal adduction torque (% weightxheight) Shoulder horizontal adduction torque (Nm) Elbow varus torque* (% weightxheight) Elbow varus torque (Nm)
3.5 (1.0) 43.7 (17.7) 172.4 (7.7) 3.9 (I.0) 49,6 (29,2) 3,9 (0.9) 47,4 (17.0)
5.0 57.8 165.3 3.5 41.4 5.3 60.2
(1.1) (14.7) (8.1) (2.5) (18.6) (1.1) (12,9)
Peak values Shoulder internal rotation torque* (% weightxheight) Shoulder internal rotation torque* (Nm) Time prior to impact (s) Shoulder horizontal adduction torque (%weightxheight) Shoulder horizontal adduction (Nm) Shoulder anterior force (% weight) Shoulder anterior force (N) Shoulder proximal force (% weight) Shoulder proximal force (N) Elbow flexion torque* (% weightxheight) Elbow flexion torque (Nm) Elbow varus torque* (% weightxheight) Elbow varus torque (Nm)
4.3 (1.0) 55.6 (18.0) 0.01 (0.02) 6.9 (1.0) 85,2 (26.4) 34.1 (10.7) 238,7 (108.8) 66.2 (12.3) 458.9 ( 1 6 2 . 3 ) 1.7 (1.3) 20,4 (19.6) 5.2 (0.8) 62.7 (14.3)
5.6 (1.2) 63.9 (12.4) 0.01 (0.01) 6.9 (1.7) 81,8 (26,4) 32,7 (16.1) 205.1 (88.9) 69.9 (13.7) 467,7(150.5) 3.1 (1.3) 36.0 (15.8) 6.3 (0.6) 73.9 (15.2)
* Significant difference at the 0.01 level. Table 2:
Mean(Standard Deviation) data from players with an effective and minimal leg-flexion..
Nm) to t h a t reported b y B a h a m o n d e (11) for lesser level payers. A significantly higher p e a k v a r u s torque w a s also recorded by the group who flexed their front knee b y 6 ° at MER (6.3%: 73.9 Nm}(F=6.0, p=0.01), w h e n c o m p a r e d with the group with a m e a n of approximately 15 ° of front knee flexion {5.2%: 62.7 Nm). Interestingly the group with the m o r e effective knee flexion recorded a lesser level t h a n the p e a k 74 Nm reported b y B a h a m o n d e (11) for male college players. A significantly reduced normalised elbow flexion torque w a s also recorded for those players with more effective knee flexion (1.7%) w h e n c o m p a r e d with the less effective group (3.1%). O
Full v e r s u s a b b r e v i a t e d b a c k s w i n g
Those players who served with a modified service action p r o d u c e d similar shoulder torques at MER and for the m e a n p e a k values, w h e n c o m p a r e d with those who u s e d a full swing. While normalised and absolute forces at the shoulder were similar between groups, there was a trend of a higher normalised anterior force at the shoulder joint for those players with a n abbreviated swing (40.4%) c o m p a r e d with those players who u s e d a full backswing (30. l%)(t= 2.34, p=0.05). Players with a n abbreviated swing there-
84
Technique effects on upper limb loading in the tennis serve
Anthropometric and ball velocity data
Full (SF & 5M)
Abreviated (4F & 3M)
Height (m) Mass (kg) Velocity (km hr 1)
179.4 (9.4) 69.4 (12.6) 163.8 (22.0)
176.2 (11.1) 66.5 (9.8) 160.6 (22.6)
Data at MER Shoulder internal rotation torque (% weightxheight) Shoulder internal rotation torque (Nm) Shoulder external rotation angle (°) Shoulder horizontal adduction torque (% weightxheight) Shoulder horizontal adduction torque (Nm) Elbow varus torque (%weightxheight) Elbow varus torque (Nm)
4,2 (1,2) 52.2 (20.8) 171.5 (8.7) 3.9 (1.6) 48.9 (27,0) 4.4 (0.9) 54.5 (18.4)
3,5 41,3 167,4 3.6 42.6 4.0 46.9
Peak values Shoulder internal rotation torque (% weightxheight) Shoulder internal rotation torque (Nm) Shoulder horizontal adduction torque (% weightxheight) Shoulder horizontal adduction (Nm) Shoulder anterior force** (% weight) Shoulder anterior force (N) Shoulder proximal force (% weight) Shoulder proximal force (N) Elbow flexion torque (% weightxheight) Elbow flexion torque (Nm) Elbow varus torque (%weightxheight) Elbow varus torque (Nm)
4.7 (1.0) 57.1 (19,8) 6.9 (1,1) 86.1 (30,0) 30,1 (13.0) 207,1 (99.8) 67.0 (13.3) 465.0 (163.1) 1.7 (1.5) 22.5 (22.3) 5,4 (1,0) 65.8 (19.0)
4,8 (1,5) 57.2 (20.0) 6.9 (1.1) 81,1 (22,2) 40,4 (6,8) 267.5 (75.9) 67.9 (11.8) 455.1 (150,7) 2.7 (1.2) 31.0 (14.9) 5.8 (0.8) 66.9 (9.3)
(1,3) (18,6) (10,1) (1.6) (24.7) (1.5) (21.5)
** Significant difference at the 0.05 level as calculated using a t-test. Table 3:
Mean (Standard Deviation) data from players with a full and abbreviated backswing.
fore h a d m o r e force applied by the t r u n k to the u p p e r a r m in a forward direction. These forces were generally smaller t h a n reported b y Noffal {12) However, Noffal (12) included ball i m p a c t in his calculations, which K n u d s o n a n d B a h a m o n d e (25) suggested m a y alter the p e a k values. Similar elbow v a r u s a n d flexion torques were recorded for b o t h "swing types" at MER a n d during the swing of the r a c k e t to impact. The s a m p l e size a n d homogeneity of the d a t a from the m a l e and female players reduced the probability of recording significant differences.
Conclusions Dillman et al. (la) stated t h a t a n y torque greater t h a n 50 Nm in the u p p e r extremity w a s a significant factor in loading t h a t a r e a of the body. This level was achieved for shoulder internal rotation torque, shoulder horizontal adduction torque, and elbow v a r u s torque during different p h a s e s of the service action. Force and torque levels at the shoulder and elbow joints were often similar to those recorded for high p e r f o r m a n c e baseball pitchers. It is therefore a p p a r e n t t h a t the shoulder a n d elbow joints are subject to high loads during the service action, and this m o v e m e n t if repeated m a n y times, m a y have
85
Technique effects on upper limb loading in the tennis serve
the potential to cause an overuse injury (26). Training m u s t be s u c h that the muscles surrounding these joints are strengthened both in eccentric and concentric movement patterns to help protect the region from injury. However, it should be stressed that physical preparation m u s t encompass all sections of the body that play a role in the kinetic chain (14, 15). Male players commonly recorded higher torques and forces at the shoulder and elbow joints t h a n their female counterparts. These higher kinetic measures are a n important factor in producing the significantly higher service velocity for this group of players. Subjects with a smaller front knee flexion and t h u s lesser "leg-drive" also loaded the shoulder and elbow joints with larger torques, particularly at MER. Players should therefore develop an effective "legdrive" during the service action to attain high velocities with as small a loading profile as possible. This requires that players flex the knee joint during the backswing phase of the service action, prior to rapidly extending during the drive to impact. While those with an abbreviated swing recorded a similar service velocity and u p p e r limb torques to those who used a full swing, minor differences were recorded in the force profile at the shoulder. A trend of higher anterior shoulder force and the general trend of forces and torques for both groups would indicate that the full swing may be a preferred service technique from a loading perspective.
Acknowledgement The authors t h a n k The Pfizer-IOC Olympic Applied Sport Research program and the International Tennis Federation for their support of this research.
References 1. Elliott, B. C., Marsh, T. & Blanksby, B. A three-dimensional cinematographical analysis of the tennis serve. Int J Sport Biom 1986; 2(4): 260-270. 2. Elliott, B. C., Marshall, R. N. & Noffal, G. J. Contributions of upper limb segment rotations during the power serve in tennis. J Appl Biomech 1995; 11 (4): 433-442. 3. Whiting, W. & Zernicke, R. Biomechanics of Musculoskeletal Injury. H u m a n Kinetics. Champaign, USA. 1998. 4. Hang, Y. S. & Peng, S. M. An epidemiologic study of upper extremity injury in tennis players with a particular reference to tennis elbow. J Form Med Assoc 1984; 83: 307-317. 5. Saal, J. Tennis, In R. Watkins (Ed.). The Spine in Sports. Mosby. New York, 1996. 6. Hill, J. Epidemiologic perspective on shoulder injuries. Clin Sport Med 1983; 2(2): 241-246. 7. Andrews, J. R. & Wilk, K. E. Shoulder injuries in baseball, In J. R. Andrews & K. E. Wilk (Eds), The Athlete's Shoulder. Churchill Livingstone. New York, 1994. p. 369-389. 8. McCann, P. D. & Bigliani, L. U. Shoulder pain in tennis players. Sports Med 1994, 17(1): 53-64. 9. Harding, W.G. Use and misuse of the tennis elbow strap. Physician Sports Med 1992; 20(8):65-74. 10. Fleisig G., Barrentine S., Zheng, N., Escamilla, R. & Andrews, J. Kinematic and kinetic comparison of baseball pitching among various levels of development. J Biom 1999; 32 (12): 1371-1375. 11. Bahamonde, R. I~. Kinetic analysis of the serving arm during the performance of the tennis serve, In R. J. Gregor, R. F. Zernicke, & W. C. Whiting (Eds.), XIIth International Society of Biomechanics (abstract #99). Los Angeles: University of California. 1989. 12. Noffal, G. Where do high speed tennis serves come from? In B. Elliott, B. Gibson & D. Knudson (Eds), Applied Proceedings oJ~the XVII International Symposium on Biomechanics in Sports: Tennis Edith Cowan University Press. Perth, Australia. 1999. p. 27-34 13. Dillman, C. J., Schultheis, J. M., Hintermeister R. A. & Hawkins R. J. What do we know about body mechanics involved in tennis skills? In H. Krahl, H Pieper, B. Kibler & P. Renstrom (Eds), Tennis: Sports Medicine and Science, Society for Tennis Medicine and Science. Auglage, 1995. p.6-11.
86
Technique effects on upper limb loading in the tennis serve 14. Kibler, W. B. Current concepts for shoulder biomechanics for tennis, In H. Krahl, H Pieper, B. Kibler & P. Renstrom (Eds), Tennis: Sports Medicine and Science, Society for Tennis Medicine and Science. Auglage. 1995. p. 59-71. 15. Kibler, W. B. Biomechanical analysis of the shoulder during tennis activities, Clin Sport Med 1995; 14(1): 79-85. 16. Fleisig, G. S., Andrews, J. R., Dillman, C. J. & Escamilla, R. F. (1995). Kinetics of baseball pitching with implications about injury mechanisms. A m J Sport Med; 23(2): 233-239. 17. Elliott, B. & Saviano, N. Serves and returns, In P. Roetert & J. Groppel (Eds) World-Class Tennis Technique., Human Kinetics. Champaign. 2001.p 207-222. 18. Feltner, M. & Dapena, J. Dynamics of the shoulder and elbow joints of the throwing arm during a baseball pitch. Int J Sport Biom 1986; 2: 235-259. 19. USRA. Racquet Tech 2000. 25(4), 4-17. 20. Brody, H. The moment of inertia of a tennis racket. The Physics Teacher 1985; 23(4): 213216. 21. Clauser, C. E., McConville, J. T. & Young, J. W. Weight, volume, and center of m a s s of segments of the h u m a n body. AMRL technical Report. Wright-Patterson Air Force Base, Ohio 1969. 22. Dempster, W. T. Space requirements of the seated operator. AMRL technical Report. Wright-Patterson Air Force Base, Ohio. 1955. 23. Dapena J. A method to determine the angular m o m e n t u m of a h u m a n body about three orthogonal axes passing through its center of gravity. J Biom 1978; 11:251-256. 24. Thomas, J. & Nelson, J. Research Methods in Physical Activity, (4th Edn). H u m a n Kinetics. Champaign, USA, 2001. 25. Knudson, D. & Bahamonde, R. Effect of endpoint conditions on position and velocity near impact in tennis, J Sport Sci 2001; 19(11):839-844. 26. Renstrom, P. & Johnson, J. Overuse injuries in sport, Sports Med 1995; 2:316-333.
87
Elliott, B, Fleisig, G, Nicholls, R, & EscamiUa, R (2003). Technique effects on upper limb loading in the tennis serve. Journal of Science and Medicine in Sport 6 (1): 76-87. The purpose of this study was to compare the shoulder and elbow joint loads during the tennis serve. Two synchronised 200 Hz video cameras were used to record the service action of 20 male and female players at the Sydney 2000 Olympics. The displacement histories of 20 selected landmarks, were calculated using the direct linear transformation approach. Ball speed was recorded from the stadium radar gun. The Peak Motus system was used to smooth displacements, while a customised inverse-dynamics program was used to calculate 3D shoulder and elbow joint kinematics and kinetics. Male players, who recorded significantly higher service speeds (male=183 km hr-l: female=149 km hr -1) recorded significantly higher normalised and absolute internal rotation shoulder torque at the position when the arm was maximally externally rotated (MER) (male=4.6% and 64.9 Nm: female=3.5% and 37.5 Nm). A higher absolute elbow varus torque (67.6 Nm) was also recorded at MER, when compared with the female players (41.3 Nm). Peak normalised horizontal adduction torque (male=7.6%: female=6.5%), normalised shoulder compressive force (male=79.6%: female=59.1%) and absolute compressive force (male=608.3 N: female=363.7 N}, were higher for the male players. Players who flexed at the front knee by 7.6 °, in the backswing phase of the serve, recorded a similar speed (162 km hr-1), and an increased normalised internal rotation torque at MER (5.0%), when compared with those who flexed by 14.7 °. They also recorded a larger normalised varus torque at MER (5,3% v 3.9%) and peak value (6.3% v 5.2%). Players who recorded a larger knee flexion also recorded less normalised and absolute (4.3%, 55.6 Nm) peak internal rotation torque compared with those with less flexion {5.6%, 63.9 Nm). Those players who used an abbreviated backswing were able to serve with a similar speed and recorded similar kinetic values. Loading on the shoulder and elbow joints is higher for the male than female players, which is a reason for the significantly higher service speed by the males, The higher kinetic measures for the group with the lower knee flexion means that all players should be encouraged to flex their knees during the backswing phase of the service action. The type of backswing was shown to have minimal influence on service velocity or loading of the shoulder and elbow joints.
Introduction T h r e e - d i m e n s i o n a l (3D) k i n e m a t i c a n a l y s e s of t h e service t e c h n i q u e of h i g h performance tennis players have provided the athlete a n d coach with practical i n f o r m a t i o n o n t h e key m e c h a n i c a l c h a r a c t e r i s t i c s of t h i s a c t i o n (1,2). However, s p o r t p h y s i c i a n s , t h e r a p i s t s , a n d t r a i n e r s n e e d to k n o w t h e forces a s s o c i a t e d
76
Technique effects on upper limb loading in the tennis serve
with these movements, as the aetiotogy of musculoskeletal injuries generally h a s a kinetic mechanical c a u s e (a). A better u n d e r s t a n d i n g of the physical d e m a n d s placed u p o n the body during this activity will enable health professionals to design more optimal injury prevention, t r e a t m e n t a n d rehabilitation programs. While epidemiological studies have reported a range of injuries in tennis, elite players c o m m o n l y suffer from shoulder a n d medial elbow injuries (4, 5). Hill (6} reported 56% of competitive players suffered rotator cuff and i m p i n g e m e n t injuries to the shoulder. It h a s b e e n h y p o t h e s i s e d t h a t the combination of a b d u c t i o n a n d external rotation of the u p p e r a r m overloads the static a n d dynamic stabilisers of the shoulder joint (7, 8). A high elbow v a r u s force is also p r e s e n t during the above actions, which h a s b e e n identified as the tennis stroke t h a t Causes m o s t medial elbow pain (9). In addition to these p r e p a r a t o r y m o v e m e n t s , the u p p e r a r m is involved in explosive inward rotation prior to ball impact, which m a y place additional s t r e s s on the shoulder region (2). While it h a s b e e n well d o c u m e n t e d t h a t the action of throwing in baseball is associated with high loads at the s h o u l d e r and elbow joints (1o), there is a paucity of information describing the loading of the shoulder and elbow joints during the tennis serve. B a h a m o n d e (111 a n d Noffal (12) in data presented as conference abstracts, reported p e a k shoulder internal rotation torques of approximately 50 Nm a n d 100 Nm respectively as the t r u n k rotates forward a n d the s h o u l d e r joint externally rotates. This occurs at the extreme limits of the shoulder joint range of motion, during the late stage of the service backswing a n d early forwardswing. They also reported a p e a k v a r u s torque at the elbow of approximately 43 Nm a n d 106 Nm respectively at this p h a s e of the service action. The u p p e r a r m in abducting, horizontally flexing a n d vigorously inwardly rotating to ball i m p a c t p r o d u c e d large shoulder horizontal flexion (164 Nm) a n d elbow v a r u s (74 Nm) torques (11). Dillman et al. (la) stated t h a t a n y torque greater t h a n 50 Nm in the u p p e r extremity was a significant factor in loading t h a t a r e a of the body. It is therefore a p p a r e n t that the u p p e r limb is subject to high loads during the service action, a n d this m o v e m e n t if repeated m a n y times m a y have the potential to c a u s e injury. Kibler (14, is) indicated t h a t a n y disruption to the kinetic chain could result in increased loading of other joints in the sequence of movements. While a coordinated action m a y reduce u p p e r limb loading in the serve, it is plausible t h a t modifications to this flow m a y increase forces on the shoulder and elbow joints. In baseball, changes to technique have been advocated in a n a t t e m p t to reduce forces a n d torques at the shoulder a n d elbow joints (16). Saal {5)stressed the need for a "deep knee bend" in the p r e p a r a t o r y tennis service m o v e m e n t s to avoid t h o r a c o l u m b a r hyperextension. Both a full backswing and a n effective "leg-drive" have b e e n advocated as features of a n efficient service technique (17) No r e s e a r c h h a s been published t h a t shows how modification to these actions affects shoulder and elbow joint loading. The p u r p o s e of this s t u d y w a s to c o m p a r e two variations in service technique for elite players u n d e r c h a m p i o n s h i p m a t c h conditions. Shoulder and elbow joint torques a n d forces at key p h a s e s of the service action were c o m p a r e d for players with a full and a n abbreviated backswing, a n d for those with a larger front knee joint flexion c o m p a r e d to those with minimal knee flexion. The above kinetic variables were also c o m p a r e d for male and female players. 77
Technique effects on upper limb loading in the tennis serve
Methods and Procedure All data were collected during singles matches on centre court during the XXVII Olympic Games in Sydney, Australia. Since professional tennis players are allowed in Olympic competition, most of the competitors were of professional status. Thirty-six tennis m a t c h e s were videotaped over a seven-day period with two electronically synchronised high-speed video cameras. H u m a n rights clearance was obtained from universities involved in the project and from the International Olympic Committee and the International Tennis Federation. Video data were collected at a rate of 200 Hz. For each camera, the s h u t t e r rate was set at 0.001 s and the aperture was adjusted according to weather conditions. Each camera was positioned approximately 20-40 m from the service area, with both side a n d front views. These views were adjusted to capture the entire serving motion, while optimising the field of view (Figure 1). Ball velocity was recorded as the ball left the tennis racket from a r a d a r g u n positioned in-line with the c e n t r e - m a r k and perpendicular to the baseline at the opposite end of the stadium. A 2 x l . 9 x l . 6 m calibration frame (Peak Performance Technologies, Inc., Englewood, CO), surveyed with a m e a s u r e m e n t tolerance of 0.005m, was videotaped at one end of the court prior to and after each match. The position of the frame e n c o m p a s s e d the space used by players during the serve (from initial ball toss to impact) on the ad-side (left-side) of one end of the court. Since kinematic parameters were m e a s u r e d only from ball toss to j u s t prior to ball impact, and not during the follow-through phase, the volume was large enough to contain the portion of the tennis serve that was digitised.
camer~
!!ii i
camerE
Figure 1: Camera locations on the centre court.
78
Technique effects on upper limb loading in the tennis serve
All videotapes were qualitatively analysed, a n d it was determined t h a t d a t a from 20 players were of the quality required for quantitative analysis. For each of these subjects, the three successful serves with the greatest ball velocity were m a n u a l l y digitised with the Peak Motus s y s t e m (Peak Performance Technologies, Inc., Englewood, CO). A 20-point spatial model was created, comprising the centres of the left a n d right mid-toes, ankles, knees, hips, shoulders, elbows, wrists, and the subject's head, h a n d racket grip, and four points a r o u n d the racket h e a d (medial, lateral, proximal, and distal). E a c h of these 20 points was digitised in every video field (200 Hz), starting at five f r a m e s prior to w h e n the ball left the h a n d a n d ending one frame prior to the racket impacting the ball. The Peak Motus s y s t e m t h e n calculated the 3D coordinate d a t a from the 2D digitised images utilising the direct linear t r a n s f o r m a t i o n method. An average root m e a n s q u a r e calibration error of 0.014 m was produced. A c o m p u t e r p r o g r a m was specifically written to calculate kinematic and kinetic p a r a m e t e r s . In this program, 3D coordinate d a t a were filtered with a 12 Hz low-pass Butterworth filter. Shoulder external rotation was calculated as the angle between the forearm and the anterior direction of the hittingshoulder, in a plane perpendicular to the long-axis of the u p p e r a r m (a horizontal f o r e a r m positioned anterior to the shoulder r e p r e s e n t s 0 °, while 90 ° of rotation h a s the forearm vertically aligned w h e n the u p p e r a r m is a b d u c t e d 90 ° to the trunk). Because external rotation is calculated using the forearm instead of the u p p e r arm, the a c c u r a c y of the calculation m a y diminish as the forearm a n d u p p e r a r m s e g m e n t s b e c o m e parallel. No m e a s u r e s were t a k e n at impact, w h e n this alignment is m o s t likely to occur. Flexion of the lead k n e e w a s determined as the relative angle between the thigh and the s h a n k (full knee extension equals 0 ° of flexion). Resultant force and torque levels at the elbow and shoulder joints were calculated u s i n g inverse dynamics. B e c a u s e of the dynamic effect of i m p a c t on the racket, kinetic analysis was limited to d a t a j u s t before ball impact. Drag effect of e a c h s e g m e n t t h r o u g h the air w a s neglected. Shoulder force w a s reported as the p r o x i m a l / d i s t a l and a n t e r i o r / p o s t e r i o r c o m p o n e n t s of the force applied by the t r u n k onto the u p p e r a r m (Figure 2A). Following a procedure first described b y Feltner and D a p e n a (is}, the s u m of all torques applied to each segment w a s set equal to the vector p r o d u c t of the s e g m e n t ' s m o m e n t of inertia a n d a n g u l a r acceleration. Moment of inertia of the racket a b o u t its medial-lateral axis w a s c o m p u t e d using the parallel axis t h e o r e m a n d published racket "swingweight" d a t a (19). Racket m o m e n t of inertia a b o u t the long-axis was calculated as: Moment of inertia (kgom 2) = Mass (kg)* [head width (m)]2/17.75 (20). Racket m o m e n t of inertia a b o u t its anteriorposterior axis w a s the s u m of the racket's other two principal m o m e n t s of inertia (20). Moment of inertia of the h a n d was a s s u m e d to be negligible. The positions of centre of m a s s of the f o r e a r m and u p p e r a r m were determined using cadaveric data (21). The m a s s of each u p p e r extremity s e g m e n t w a s a s s u m e d to be a percentage of the subject's total m a s s {211. Moments of inertia of the forearm a n d u p p e r a r m were initially determined from D e m p s t e r ' s (22) cadaveric study, and t h e n individualised using each subject's height a n d m a s s (2a). Shoulder internal rotation torque and horizontal adduction torque were the c o m p o n e n t s of the torque applied b y the t r u n k onto the u p p e r a r m
79
Technique effects on upper limb loading in the tennis serve
Anterior
\ Figure 2: Convention for kinetic measurements: (A) shoulder force, (B) shoulder torque, (C) elbow torque
Figure 3: Front knee joint flexion (~) in the service backswing (straight limb = 0 ° flexion).
Figure 4: Full (A) and abbreviated (B) backswings in the serve.
8O
Technique effects on upper limb loading in the tennis serve /
(Figure 2B). Elbow v a r u s torque was reported as the torque applied by the u p p e r a r m onto the forearm a b o u t the anterior direction of the f o r e a r m (Figure 2C). Elbow flexion torque was the torque applied b y the u p p e r a r m onto the forearm, a b o u t the medial direction of the elbow. Repeatability of data was a s s e s s e d by redigitising 30 consecutive frames, after one week, for the period from MER to immediately before impact. Shoulder internal rotation a n d elbow extension angles were c o m p a r e d using S t u d e n t ' s t-tests. Mean d a t a re-digitised over the 30 consecutive frames were shown to be reliable in t h a t shoulder internal rotation and elbow extension were not significantly different and varied by 0.8% a n d 0.4% respectively. Players were classified as exhibiting a n effective (>10 ° of knee flexion; Figure 3) or less-effective leg drive (<10 ° of knee flexion); at MER in the service action. Players were also categorised by their type of backswing. Players who u s e d a full rotation of the shoulder joint during the service action (Figure 4A) were classified as full backswing, w h e r e a s players who lifted the racket vertically during the early backswing (Figure 4B) were considered to have an abbreviated backswing. Players were also classified b y their gender. H y p o t h e s e s were that m a l e s would record higher kinetic values t h a n females, a n d those players with a n effective leg drive would record lower values t h a n t h o s e w i t h m i n i m a l k n e e flexion d u r i n g t h e b a c k s w i n g . It w a s also h y p o t h e s i s e d t h a t players with a full b a c k s w i n g would record lower kinetic values t h a n those with a n abbreviated backswing. All c o m p a r i s o n s were tested t h r o u g h a three way ANOVA (Data d e s k 6.1: D a t a Description Inc., USA). Ttests for u n e v e n cell sizes were u s e d to f u r t h e r c o m p a r e data where trends were identified from the above statistics (24). A 0.01 level of acceptance was established with respect to the above hypotheses. In addition to absolute values, n o r m a l i s e d values of all force and torque d a t a were calculated to improve c o m p a r i s o n of groups containing varying n u m b e r s of male a n d female players. Force d a t a were divided b y bodyweight and t h e n multiplied b y 100%, as h a s occurred in the baseball literature (16). Torque d a t a were divided by weight by height, t h e n multiplied b y 100% (16).
Results and DiScussion Gender, age, body mass, height, and service velocity for the m e a n of the b e s t three trials are s h o w n in Table 1. Serving speeds for the male players (182.8 k m h r -1) w a s significantly higher (F=24.7, p=0.01) t h a n for the female competitors (149.3 k m hr-]). The speed recorded b y both groups testifies to the elite quality of this sample, as values for high p e r f o r m a n c e d a t a in the literature are reported as approximately 155 a n d 125 k m hr -1 for male a n d female players respectively (1).
Male and female players Although the m e a n angle at MER w a s similar between groups (~170°), significantly larger normalised (F=5.4, p<0.01) a n d absolute (F= 12.2, p<0.01) internal rotation torques were recorded at this i n s t a n t for male (4.6%: 64.9 Nm) c o m p a r e d with female competitors (3.5%: 37.5 Nm) (Table 1). Similar m e a n angle (165 °) a n d absolute internal rotation torque (67 Nm) were recorded at MER for highly skilled male pitchers (16). While normalised m e a n elbow v a r u s torque at MER w a s similar for both groups, the absolute value was significantly higher for male players (67.6 Nm) t h a n for their female c o u n t e r p a r t s (41.3 Nm)
81
Technique effects on upper limb loading in the tennis serve
Anthropometric and ball velocity data
Males (n=8)
Females (n=12)
Height* (m) Mass* (kg) Velocity* (km hr1)
185.0 (5.5) 81.0 (9.0) 182.8 (13,9)
174.0 (8.6) 62.2 (7.7) 149.3 (14,0)
Data at MER Shoulder internal rotation torque* (% weightxheight) Shoulder internal rotation torque* (Nm) Shoulder external rotation angle (°) Shoulder horizontal adduction torque (% weightxheight) Shoulder horizontal adduction torque (Nm) Elbow varus torque (% weightxheight) EIbowvarus torque* (Nm)
4.6 (0.9) 64.9 (15,8) 169,2 (9,0) 4,2 (1,7) 61.7 (31.0) 4.8 (0.9) 67.6 (15,2)
3,5 (1,2) 37,5 (15.0) 170.9 (8.1) 3.5 (1,4) 36,8 (16,3) 3,9 (1.2) 41.3 (14,1)
Peak values Shoulder internal rotation torque (% weightxheight) Shoulder internal rotation torque (Nm) Time prior to impact (s) Shoulder horizontal adduction torque* (% weightxheight) Shoulder horizontal adduction (Nm) Shoulder anterior force (% weight) Shoulder anterior force (N) Shoulder proximal force* (%weight) Shoulder proximal force* (N) Elbow flexion torque (%weightxheight) Elbow flexion torque (Nm) Elbow varus torque (% weightxheight) EIbowvarus torque** (Nm)
5.1 (0,9) 71.2 (15,1) 0.06 (0.02) 7.6 (0,8) 107.8 (24,9) 38.5 (14,0) 291.7 (119.8) 79.6 (5.3) 608.3 (109,5) 2.6 (1.6) 36.7 (22.9) 5.6 (0.9) 78.3 (12.2)
4.5 (1.3) 47.8 (16,3) 0,06 (0,02) 6,5 (0,9) 68,8 (14.3) 30.5 (10.2) 185.1 (60.9) 59.1 (8.4) 363.7 (87,8) 1,7 (1,3) 18.1 (14.5) 5.3 (1,1) 58.2 (13.1)
* Significant difference at the 0.01 level. **Significant difference at the 0.05 level as calculated using a t-test. Table 1:
Mean (Standard Deviation) service data for male and female players.
(F=13.2, p<0.01). Again data from the male players were almost identical to the 64 Nm reported by Fleisig et al. (16) for male baseball pitchers. The peak normalised horizontal adductor torque was significantly higher for the male players (7.6%)(F=7.47, p<0.01) when compared with the females (6.5%), while other shoulder torque values were similar in the swing to impact (Table 1). The m e a n horizontal adduction torque (107.8 Nm) for the male players was less t h a n reported by B a h a m o n d e (11) for college-level male tennis players (164 Nm), who p r e s u m a b l y served at a lesser speed t h a n this sample of professional players. Mean peak internal rotation torque values were relatively high during this phase of the service action for both groups (males = 71.2 Nm and females = 47.8 Nm), which supports the importance of internal rotation as one of the most important contributors to racket speed at impact (2). The level of internal rotation torque for the male players was larger t h a n the 33 Nm reported j u s t prior to impact by Bahamonde (11). However, it should be stressed that the values reported in this study were peak values, and not those immediately prior to impact. Normalised (F=6.7, (proximal) p<0.01) and absolute (F=7.3, p<0.01) m e a n peak shoulder compressive proximal forces were significantly larger for the 82
Technique effects on upper limb loading in the tennis serve
male players (79.6%: 608.3 N) c o m p a r e d with the female competitors (59.1%: 363.7 N). This is again similar to baseball (16). To prevent distraction at the shoulder d u n n g this p a r t of the action, the t r u n k applies a r e s u l t a n t force to the u p p e r a r m to stabilise the joint. Anterior forces a b o u t the shoulder joint were similar for all players irrespective of sex. The m a g n i t u d e s of shoulder kinetics during the p r e s e n t tennis s t u d y for the male players were smaller t h a n the shoulder kinetics previously reported for baseball pitching (16, 10) There w a s a trend for p e a k absolute elbow v a r u s torque to be higher for m a l e s (78.3 Nm) w h e n c o m p a r e d with the female players (58.2 Nm). The torque m a g n i t u d e for the male players was greater t h a n the 64 Nm reported for highly skilled a n d professional male baseball pitchers b y Fleisig et al. {16, 10), b u t less t h a n the 100 N m recorded by Feltner a n d D a p e n a t181 for baseball pitching. The fact t h a t the male players served at a significantly higher m e a n velocity was p r e s u m a b l y the result of the higher kinetic values in the u p p e r limb. The relationship between service velocity a n d u p p e r limb kinetics, particularly with respect to injury, is the subject of further investigation. Level Of knee flexion At MER there was a significantly lower n o r m a l i s e d internal rotation torque (3.5%) recorded by the group with a m e a n knee flexion of 15.9 °, w h e n c o m p a r e d with the group with a m e a n knee flexion 5.8 ° (5.0%: F= 8.08, p= 0.01). The group with the larger flexion recorded a similar level (43.7 Nm), a n d the less effective group a greater level (57.8 Nm) t h a n the 43 Nm reported by B a h a m o n d e (1~), for college level tennis players. However, both groups recorded lower values t h a n the 94 Nm reported by Noffal {12). This torque was associated with the eccentric contraction of the m u s c l e s responsible for internal shoulder rotation (eg latissimus dorsi, pectoralis m a j o r and subscapularis, teres major, and anterior deltoid) following MER. The players with the potential for a better leg-drive m a y therefore u s e the inertial transfer from the t r u n k to u p p e r limb to move the u p p e r a r m into a position of MER. This t h e n requires less internal rotator torque to stop the external rotation. The group with less effective drive m u s t therefore primarily use the external rotators to achieve MER, which t h e n requires a greater internal rotator torque to reverse the rotation of the u p p e r arm. A significantly reduced normalised (4.3 v 5.6%) a n d absolute p e a k internal rotation torque (55.6 Nm v 63.9 Nm), w a s recorded for the group with the larger knee joint flexion, w h e n c o m p a r e d with the less effective leg-flexion group. This reduction placed a lesser load on the joint during the concentric contraction of the m u s c l e s involved in rotating the u p p e r a r m in the swing to impact. The group with the larger knee flexion recorded a lower m e a n m a x i m u m internal rotator torque a n d the group with the lesser flexion a similar torque to the 68 Nm reported for professional pitchers (lo). Lesser load is t h u s evident with the group with the larger front knee flexion in performing an action described by Elliott et al. (2) as being integral to an effective service action. No significant differences were recorded w h e n the shoulder forces were c o m p a r e d between the two groups. A significantly larger m e a n elbow v a r u s torque was recorded at the position of MER for those players with a lesser knee flexion (5.3%: 60.2 Nm v 3.9%: 47.4 Nm). The group with the larger knee flexion recorded a similar v a r u s torque (43
83
Technique effects on upper limb loading in the tennis serve
Anthropometric and ball velocity data
Effective (9F & 5M)
Minimal (3F & 3M)
Height (m) Mass (kg) Service velocity (km hr "1) Knee flexion* (°)
178.3 (10.1) 69.0 (12.2) 162.9 (20.4) 15.9 (4.9)
178.5 66.5 162.2 5.8
(9.9) (11.0) (26.6) (1.8)
Data at MER Shoulder internal rotation torque* (% weightxheight) Shoulder internal rotation torque (Nm) Shoulder external rotation angle (°) Shoulder horizontal adduction torque (% weightxheight) Shoulder horizontal adduction torque (Nm) Elbow varus torque* (% weightxheight) Elbow varus torque (Nm)
3.5 (1.0) 43.7 (17.7) 172.4 (7.7) 3.9 (I.0) 49,6 (29,2) 3,9 (0.9) 47,4 (17.0)
5.0 57.8 165.3 3.5 41.4 5.3 60.2
(1.1) (14.7) (8.1) (2.5) (18.6) (1.1) (12,9)
Peak values Shoulder internal rotation torque* (% weightxheight) Shoulder internal rotation torque* (Nm) Time prior to impact (s) Shoulder horizontal adduction torque (%weightxheight) Shoulder horizontal adduction (Nm) Shoulder anterior force (% weight) Shoulder anterior force (N) Shoulder proximal force (% weight) Shoulder proximal force (N) Elbow flexion torque* (% weightxheight) Elbow flexion torque (Nm) Elbow varus torque* (% weightxheight) Elbow varus torque (Nm)
4.3 (1.0) 55.6 (18.0) 0.01 (0.02) 6.9 (1.0) 85,2 (26.4) 34.1 (10.7) 238,7 (108.8) 66.2 (12.3) 458.9 ( 1 6 2 . 3 ) 1.7 (1.3) 20,4 (19.6) 5.2 (0.8) 62.7 (14.3)
5.6 (1.2) 63.9 (12.4) 0.01 (0.01) 6.9 (1.7) 81,8 (26,4) 32,7 (16.1) 205.1 (88.9) 69.9 (13.7) 467,7(150.5) 3.1 (1.3) 36.0 (15.8) 6.3 (0.6) 73.9 (15.2)
* Significant difference at the 0.01 level. Table 2:
Mean(Standard Deviation) data from players with an effective and minimal leg-flexion..
Nm) to t h a t reported b y B a h a m o n d e (11) for lesser level payers. A significantly higher p e a k v a r u s torque w a s also recorded by the group who flexed their front knee b y 6 ° at MER (6.3%: 73.9 Nm}(F=6.0, p=0.01), w h e n c o m p a r e d with the group with a m e a n of approximately 15 ° of front knee flexion {5.2%: 62.7 Nm). Interestingly the group with the m o r e effective knee flexion recorded a lesser level t h a n the p e a k 74 Nm reported b y B a h a m o n d e (11) for male college players. A significantly reduced normalised elbow flexion torque w a s also recorded for those players with more effective knee flexion (1.7%) w h e n c o m p a r e d with the less effective group (3.1%). O
Full v e r s u s a b b r e v i a t e d b a c k s w i n g
Those players who served with a modified service action p r o d u c e d similar shoulder torques at MER and for the m e a n p e a k values, w h e n c o m p a r e d with those who u s e d a full swing. While normalised and absolute forces at the shoulder were similar between groups, there was a trend of a higher normalised anterior force at the shoulder joint for those players with a n abbreviated swing (40.4%) c o m p a r e d with those players who u s e d a full backswing (30. l%)(t= 2.34, p=0.05). Players with a n abbreviated swing there-
84
Technique effects on upper limb loading in the tennis serve
Anthropometric and ball velocity data
Full (SF & 5M)
Abreviated (4F & 3M)
Height (m) Mass (kg) Velocity (km hr 1)
179.4 (9.4) 69.4 (12.6) 163.8 (22.0)
176.2 (11.1) 66.5 (9.8) 160.6 (22.6)
Data at MER Shoulder internal rotation torque (% weightxheight) Shoulder internal rotation torque (Nm) Shoulder external rotation angle (°) Shoulder horizontal adduction torque (% weightxheight) Shoulder horizontal adduction torque (Nm) Elbow varus torque (%weightxheight) Elbow varus torque (Nm)
4,2 (1,2) 52.2 (20.8) 171.5 (8.7) 3.9 (1.6) 48.9 (27,0) 4.4 (0.9) 54.5 (18.4)
3,5 41,3 167,4 3.6 42.6 4.0 46.9
Peak values Shoulder internal rotation torque (% weightxheight) Shoulder internal rotation torque (Nm) Shoulder horizontal adduction torque (% weightxheight) Shoulder horizontal adduction (Nm) Shoulder anterior force** (% weight) Shoulder anterior force (N) Shoulder proximal force (% weight) Shoulder proximal force (N) Elbow flexion torque (% weightxheight) Elbow flexion torque (Nm) Elbow varus torque (%weightxheight) Elbow varus torque (Nm)
4.7 (1.0) 57.1 (19,8) 6.9 (1,1) 86.1 (30,0) 30,1 (13.0) 207,1 (99.8) 67.0 (13.3) 465.0 (163.1) 1.7 (1.5) 22.5 (22.3) 5,4 (1,0) 65.8 (19.0)
4,8 (1,5) 57.2 (20.0) 6.9 (1.1) 81,1 (22,2) 40,4 (6,8) 267.5 (75.9) 67.9 (11.8) 455.1 (150,7) 2.7 (1.2) 31.0 (14.9) 5.8 (0.8) 66.9 (9.3)
(1,3) (18,6) (10,1) (1.6) (24.7) (1.5) (21.5)
** Significant difference at the 0.05 level as calculated using a t-test. Table 3:
Mean (Standard Deviation) data from players with a full and abbreviated backswing.
fore h a d m o r e force applied by the t r u n k to the u p p e r a r m in a forward direction. These forces were generally smaller t h a n reported b y Noffal {12) However, Noffal (12) included ball i m p a c t in his calculations, which K n u d s o n a n d B a h a m o n d e (25) suggested m a y alter the p e a k values. Similar elbow v a r u s a n d flexion torques were recorded for b o t h "swing types" at MER a n d during the swing of the r a c k e t to impact. The s a m p l e size a n d homogeneity of the d a t a from the m a l e and female players reduced the probability of recording significant differences.
Conclusions Dillman et al. (la) stated t h a t a n y torque greater t h a n 50 Nm in the u p p e r extremity w a s a significant factor in loading t h a t a r e a of the body. This level was achieved for shoulder internal rotation torque, shoulder horizontal adduction torque, and elbow v a r u s torque during different p h a s e s of the service action. Force and torque levels at the shoulder and elbow joints were often similar to those recorded for high p e r f o r m a n c e baseball pitchers. It is therefore a p p a r e n t t h a t the shoulder a n d elbow joints are subject to high loads during the service action, and this m o v e m e n t if repeated m a n y times, m a y have
85
Technique effects on upper limb loading in the tennis serve
the potential to cause an overuse injury (26). Training m u s t be s u c h that the muscles surrounding these joints are strengthened both in eccentric and concentric movement patterns to help protect the region from injury. However, it should be stressed that physical preparation m u s t encompass all sections of the body that play a role in the kinetic chain (14, 15). Male players commonly recorded higher torques and forces at the shoulder and elbow joints t h a n their female counterparts. These higher kinetic measures are a n important factor in producing the significantly higher service velocity for this group of players. Subjects with a smaller front knee flexion and t h u s lesser "leg-drive" also loaded the shoulder and elbow joints with larger torques, particularly at MER. Players should therefore develop an effective "legdrive" during the service action to attain high velocities with as small a loading profile as possible. This requires that players flex the knee joint during the backswing phase of the service action, prior to rapidly extending during the drive to impact. While those with an abbreviated swing recorded a similar service velocity and u p p e r limb torques to those who used a full swing, minor differences were recorded in the force profile at the shoulder. A trend of higher anterior shoulder force and the general trend of forces and torques for both groups would indicate that the full swing may be a preferred service technique from a loading perspective.
Acknowledgement The authors t h a n k The Pfizer-IOC Olympic Applied Sport Research program and the International Tennis Federation for their support of this research.
References 1. Elliott, B. C., Marsh, T. & Blanksby, B. A three-dimensional cinematographical analysis of the tennis serve. Int J Sport Biom 1986; 2(4): 260-270. 2. Elliott, B. C., Marshall, R. N. & Noffal, G. J. Contributions of upper limb segment rotations during the power serve in tennis. J Appl Biomech 1995; 11 (4): 433-442. 3. Whiting, W. & Zernicke, R. Biomechanics of Musculoskeletal Injury. H u m a n Kinetics. Champaign, USA. 1998. 4. Hang, Y. S. & Peng, S. M. An epidemiologic study of upper extremity injury in tennis players with a particular reference to tennis elbow. J Form Med Assoc 1984; 83: 307-317. 5. Saal, J. Tennis, In R. Watkins (Ed.). The Spine in Sports. Mosby. New York, 1996. 6. Hill, J. Epidemiologic perspective on shoulder injuries. Clin Sport Med 1983; 2(2): 241-246. 7. Andrews, J. R. & Wilk, K. E. Shoulder injuries in baseball, In J. R. Andrews & K. E. Wilk (Eds), The Athlete's Shoulder. Churchill Livingstone. New York, 1994. p. 369-389. 8. McCann, P. D. & Bigliani, L. U. Shoulder pain in tennis players. Sports Med 1994, 17(1): 53-64. 9. Harding, W.G. Use and misuse of the tennis elbow strap. Physician Sports Med 1992; 20(8):65-74. 10. Fleisig G., Barrentine S., Zheng, N., Escamilla, R. & Andrews, J. Kinematic and kinetic comparison of baseball pitching among various levels of development. J Biom 1999; 32 (12): 1371-1375. 11. Bahamonde, R. I~. Kinetic analysis of the serving arm during the performance of the tennis serve, In R. J. Gregor, R. F. Zernicke, & W. C. Whiting (Eds.), XIIth International Society of Biomechanics (abstract #99). Los Angeles: University of California. 1989. 12. Noffal, G. Where do high speed tennis serves come from? In B. Elliott, B. Gibson & D. Knudson (Eds), Applied Proceedings oJ~the XVII International Symposium on Biomechanics in Sports: Tennis Edith Cowan University Press. Perth, Australia. 1999. p. 27-34 13. Dillman, C. J., Schultheis, J. M., Hintermeister R. A. & Hawkins R. J. What do we know about body mechanics involved in tennis skills? In H. Krahl, H Pieper, B. Kibler & P. Renstrom (Eds), Tennis: Sports Medicine and Science, Society for Tennis Medicine and Science. Auglage, 1995. p.6-11.
86
Technique effects on upper limb loading in the tennis serve 14. Kibler, W. B. Current concepts for shoulder biomechanics for tennis, In H. Krahl, H Pieper, B. Kibler & P. Renstrom (Eds), Tennis: Sports Medicine and Science, Society for Tennis Medicine and Science. Auglage. 1995. p. 59-71. 15. Kibler, W. B. Biomechanical analysis of the shoulder during tennis activities, Clin Sport Med 1995; 14(1): 79-85. 16. Fleisig, G. S., Andrews, J. R., Dillman, C. J. & Escamilla, R. F. (1995). Kinetics of baseball pitching with implications about injury mechanisms. A m J Sport Med; 23(2): 233-239. 17. Elliott, B. & Saviano, N. Serves and returns, In P. Roetert & J. Groppel (Eds) World-Class Tennis Technique., Human Kinetics. Champaign. 2001.p 207-222. 18. Feltner, M. & Dapena, J. Dynamics of the shoulder and elbow joints of the throwing arm during a baseball pitch. Int J Sport Biom 1986; 2: 235-259. 19. USRA. Racquet Tech 2000. 25(4), 4-17. 20. Brody, H. The moment of inertia of a tennis racket. The Physics Teacher 1985; 23(4): 213216. 21. Clauser, C. E., McConville, J. T. & Young, J. W. Weight, volume, and center of m a s s of segments of the h u m a n body. AMRL technical Report. Wright-Patterson Air Force Base, Ohio 1969. 22. Dempster, W. T. Space requirements of the seated operator. AMRL technical Report. Wright-Patterson Air Force Base, Ohio. 1955. 23. Dapena J. A method to determine the angular m o m e n t u m of a h u m a n body about three orthogonal axes passing through its center of gravity. J Biom 1978; 11:251-256. 24. Thomas, J. & Nelson, J. Research Methods in Physical Activity, (4th Edn). H u m a n Kinetics. Champaign, USA, 2001. 25. Knudson, D. & Bahamonde, R. Effect of endpoint conditions on position and velocity near impact in tennis, J Sport Sci 2001; 19(11):839-844. 26. Renstrom, P. & Johnson, J. Overuse injuries in sport, Sports Med 1995; 2:316-333.
87